icarus112 commited on
Commit
f257325
·
verified ·
1 Parent(s): aa9fdfa

Upload v14p14/ckpt_150000.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. v14p14/ckpt_150000.json +1 -0
v14p14/ckpt_150000.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"sent": 150000, "total_steps": 2594813, "eval_results": [{"sent": 2000, "train_loss": 2.085123744664439, "train_acc": 0.4408124704770902, "eval_loss": 1.5152359620462006, "eval_acc": 0.44594594594594594, "ppl": 4.55049474422526, "w_freq": 0.001069752044629863, "w_bigram": 0.017389398230965417, "w_trigram": 0.07154396580765796, "w_skip_bi": 0.23805648573810956, "w_skip_tri": 0.465205430367037, "w_neural_phoneme": 0.3555831372188447, "w_neural_word": 0.3540519218134524, "word_vocab": 1593, "elapsed_s": 22.99993133544922, "steps_per_sec": 1380.6562957454057}, {"sent": 4000, "train_loss": 2.015406930693783, "train_acc": 0.449146110056926, "eval_loss": 1.5451346801541017, "eval_acc": 0.524390243902439, "ppl": 4.688603046282032, "w_freq": 0.003511844702708052, "w_bigram": 0.02321205769493682, "w_trigram": 0.0931775383776142, "w_skip_bi": 0.1623459701262452, "w_skip_tri": 0.4805304250923348, "w_neural_phoneme": 0.3684824684955725, "w_neural_word": 0.36741944722849984, "word_vocab": 2198, "elapsed_s": 40.84769368171692, "steps_per_sec": 1422.479331458738}, {"sent": 6000, "train_loss": 2.0513120565987895, "train_acc": 0.4426420659144689, "eval_loss": 2.385693284711908, "eval_acc": 0.37755102040816324, "ppl": 10.866593697324928, "w_freq": 0.0038737517974814202, "w_bigram": 0.042317026166645025, "w_trigram": 0.07485745911932162, "w_skip_bi": 0.16210056927329614, "w_skip_tri": 0.4859366519149216, "w_neural_phoneme": 0.35027883460856923, "w_neural_word": 0.3608633910216888, "word_vocab": 2819, "elapsed_s": 60.424288749694824, "steps_per_sec": 1444.1874584819293}, {"sent": 8000, "train_loss": 2.048645793655169, "train_acc": 0.4365337183420562, "eval_loss": 1.598692538850994, "eval_acc": 0.423728813559322, "ppl": 4.946560758578182, "w_freq": 0.00032966983389561694, "w_bigram": 0.0002418866251740584, "w_trigram": 0.1405691853149911, "w_skip_bi": 0.1781139578091503, "w_skip_tri": 0.53065518561347, "w_neural_phoneme": 0.3518465404687118, "w_neural_word": 0.35171821085490185, "word_vocab": 3459, "elapsed_s": 82.65294742584229, "steps_per_sec": 1458.0242296637243}, {"sent": 10000, "train_loss": 2.0392960710502033, "train_acc": 0.4366373172928835, "eval_loss": 1.5101141782385703, "eval_acc": 0.4852941176470588, "ppl": 4.5272476779707, "w_freq": 0.0005806880985354682, "w_bigram": 0.00650279613709746, "w_trigram": 0.05842327539161242, "w_skip_bi": 0.3587923930954613, "w_skip_tri": 0.55, "w_neural_phoneme": 0.3528922903222367, "w_neural_word": 0.35288160843142014, "word_vocab": 4067, "elapsed_s": 105.56567192077637, "steps_per_sec": 1467.5509299676955}, {"sent": 12000, "train_loss": 1.9055074628648694, "train_acc": 0.4613161614575153, "eval_loss": 1.6948638109062797, "eval_acc": 0.47368421052631576, "ppl": 5.4459042419434365, "w_freq": 0.0028678857932047954, "w_bigram": 0.01396978335657545, "w_trigram": 0.018171927977086527, "w_skip_bi": 0.4, "w_skip_tri": 0.55, "w_neural_phoneme": 0.3501947007286266, "w_neural_word": 0.35327151723764794, "word_vocab": 4455, "elapsed_s": 126.81924510002136, "steps_per_sec": 1472.6313806134503}, {"sent": 14000, "train_loss": 1.988357764454176, "train_acc": 0.44365667322606783, "eval_loss": 2.019602058013299, "eval_acc": 0.4647887323943662, "ppl": 7.535325714458626, "w_freq": 0.0012078577103194924, "w_bigram": 0.0018435259292709206, "w_trigram": 0.0021833600088577193, "w_skip_bi": 0.21387570596017932, "w_skip_tri": 0.5215356352951087, "w_neural_phoneme": 0.353403763158099, "w_neural_word": 0.35306716376230435, "word_vocab": 4946, "elapsed_s": 149.80217838287354, "steps_per_sec": 1477.3750448031014}, {"sent": 16000, "train_loss": 1.8859793216408223, "train_acc": 0.46099375900144024, "eval_loss": 1.8679316252331395, "eval_acc": 0.4805194805194805, "ppl": 6.474890040201499, "w_freq": 0.001179423340823594, "w_bigram": 0.012159731056554247, "w_trigram": 0.0169469737566637, "w_skip_bi": 0.2567323738435507, "w_skip_tri": 0.5097042745823981, "w_neural_phoneme": 0.35950514245749043, "w_neural_word": 0.3580844852439058, "word_vocab": 5338, "elapsed_s": 172.0545299053192, "steps_per_sec": 1480.0075309852539}, {"sent": 18000, "train_loss": 1.888715130293185, "train_acc": 0.4616469417099878, "eval_loss": 1.7267277988732628, "eval_acc": 0.46153846153846156, "ppl": 5.622226719529882, "w_freq": 0.00038247254425608033, "w_bigram": 0.0004838296631879124, "w_trigram": 0.00031333466511156654, "w_skip_bi": 0.38464589601824545, "w_skip_tri": 0.55, "w_neural_phoneme": 0.3507923324885186, "w_neural_word": 0.3507842421818335, "word_vocab": 5752, "elapsed_s": 196.0739209651947, "steps_per_sec": 1483.1447168928769}, {"sent": 20000, "train_loss": 1.8248699290471841, "train_acc": 0.4760111493298541, "eval_loss": 1.4147613475506724, "eval_acc": 0.5585585585585585, "ppl": 4.115504173613433, "w_freq": 0.000288660021262548, "w_bigram": 0.013234249848771156, "w_trigram": 0.04723915440803704, "w_skip_bi": 0.21981749405887852, "w_skip_tri": 0.5492620941688723, "w_neural_phoneme": 0.3508149228116177, "w_neural_word": 0.350867933031737, "word_vocab": 6098, "elapsed_s": 218.57639145851135, "steps_per_sec": 1484.74406515033}, {"sent": 22000, "train_loss": 1.859303542487796, "train_acc": 0.4664312020769184, "eval_loss": 1.9223074086263836, "eval_acc": 0.37681159420289856, "ppl": 6.836715379527306, "w_freq": 0.0004853969397382425, "w_bigram": 0.0022979376559019654, "w_trigram": 0.03038502034964991, "w_skip_bi": 0.321226650084887, "w_skip_tri": 0.55, "w_neural_phoneme": 0.3528267656705988, "w_neural_word": 0.35324112850123973, "word_vocab": 6393, "elapsed_s": 240.77224731445312, "steps_per_sec": 1485.4535935484894}, {"sent": 24000, "train_loss": 1.8658042247479767, "train_acc": 0.4647563340015116, "eval_loss": 1.571835950298054, "eval_acc": 0.4606741573033708, "ppl": 4.815481066836054, "w_freq": 0.0003947745855737373, "w_bigram": 0.0021288640814372695, "w_trigram": 0.008892850994688641, "w_skip_bi": 0.2151799741193014, "w_skip_tri": 0.5494586444588437, "w_neural_phoneme": 0.35164240574284245, "w_neural_word": 0.35154536242438766, "word_vocab": 6673, "elapsed_s": 261.119859457016, "steps_per_sec": 1486.2408428336512}, {"sent": 26000, "train_loss": 1.9037415796300843, "train_acc": 0.4523525999206131, "eval_loss": 1.9425843992588354, "eval_acc": 0.45161290322580644, "ppl": 6.976758420685662, "w_freq": 0.0006006081329066348, "w_bigram": 0.0005690353716866797, "w_trigram": 0.0013654439743136914, "w_skip_bi": 0.2776214065459763, "w_skip_tri": 0.5499870923599193, "w_neural_phoneme": 0.351135376628905, "w_neural_word": 0.35107115587949034, "word_vocab": 6984, "elapsed_s": 282.8300018310547, "steps_per_sec": 1487.9538849325568}, {"sent": 28000, "train_loss": 2.082350984559117, "train_acc": 0.4174616482744556, "eval_loss": 1.81444532469722, "eval_acc": 0.4539877300613497, "ppl": 6.137670818678332, "w_freq": 0.0004568467129081488, "w_bigram": 0.0013958726076375811, "w_trigram": 0.1025999549684254, "w_skip_bi": 0.0898673705321814, "w_skip_tri": 0.5449725158155078, "w_neural_phoneme": 0.3510743381651484, "w_neural_word": 0.3508859956298796, "word_vocab": 7648, "elapsed_s": 311.79852175712585, "steps_per_sec": 1491.2482502472724}, {"sent": 30000, "train_loss": 2.1732246622767892, "train_acc": 0.39509158067095695, "eval_loss": 1.8401622981332397, "eval_acc": 0.40397350993377484, "ppl": 6.297560260364343, "w_freq": 0.0007952189097450425, "w_bigram": 0.0036082508640084884, "w_trigram": 0.06551654636633653, "w_skip_bi": 0.08166040334936842, "w_skip_tri": 0.5358785756164435, "w_neural_phoneme": 0.3499206204808939, "w_neural_word": 0.3499055848770112, "word_vocab": 8515, "elapsed_s": 346.07197523117065, "steps_per_sec": 1495.3276689172076}, {"sent": 32000, "train_loss": 2.1405475605529576, "train_acc": 0.3960335077855372, "eval_loss": 1.631776288293821, "eval_acc": 0.49572649572649574, "ppl": 5.112948728095009, "w_freq": 0.00017931807862246708, "w_bigram": 0.0003158384002599147, "w_trigram": 0.03999761261350022, "w_skip_bi": 0.1859542926743709, "w_skip_tri": 0.5764628643819252, "w_neural_phoneme": 0.34240172448767076, "w_neural_word": 0.3423426166280084, "word_vocab": 9486, "elapsed_s": 386.3890173435211, "steps_per_sec": 1499.0281141584628}, {"sent": 34000, "train_loss": 2.1376852612836825, "train_acc": 0.39764329257909614, "eval_loss": 1.742513397956904, "eval_acc": 0.44155844155844154, "ppl": 5.711681124617898, "w_freq": 0.0001245063290761764, "w_bigram": 0.00016668010508834847, "w_trigram": 0.04930964258637601, "w_skip_bi": 0.12497648242560751, "w_skip_tri": 0.5939806870363084, "w_neural_phoneme": 0.33643750059565924, "w_neural_word": 0.3364241947052025, "word_vocab": 10546, "elapsed_s": 424.8360140323639, "steps_per_sec": 1501.6005680521293}, {"sent": 36000, "train_loss": 2.1102055080210835, "train_acc": 0.41044861853064507, "eval_loss": 1.8080696149578894, "eval_acc": 0.44339622641509435, "ppl": 6.098663293528725, "w_freq": 0.0003622337082672906, "w_bigram": 0.0010506896487168665, "w_trigram": 0.015212827623932798, "w_skip_bi": 0.11320936035960315, "w_skip_tri": 0.5952969749457497, "w_neural_phoneme": 0.3334897007652057, "w_neural_word": 0.3334532935887197, "word_vocab": 11088, "elapsed_s": 448.59895873069763, "steps_per_sec": 1499.673565679999}, {"sent": 38000, "train_loss": 2.03699839073014, "train_acc": 0.42727860374919197, "eval_loss": 1.7769149483305664, "eval_acc": 0.45348837209302323, "ppl": 5.9115906949850885, "w_freq": 0.0011511318236046078, "w_bigram": 0.0016170361109661177, "w_trigram": 0.003972050078071888, "w_skip_bi": 0.17111664119904046, "w_skip_tri": 0.5859780101711618, "w_neural_phoneme": 0.33069573161089094, "w_neural_word": 0.33092145665316447, "word_vocab": 11395, "elapsed_s": 467.5132691860199, "steps_per_sec": 1498.562813457253}, {"sent": 40000, "train_loss": 2.081252306942877, "train_acc": 0.4147263916794705, "eval_loss": 1.6168915354692983, "eval_acc": 0.5967741935483871, "ppl": 5.0374073521418365, "w_freq": 0.0008190454364046659, "w_bigram": 0.0019028183425004144, "w_trigram": 0.03132833177391257, "w_skip_bi": 0.21179981342197343, "w_skip_tri": 0.6311354334934055, "w_neural_phoneme": 0.327529391784775, "w_neural_word": 0.3274781410298536, "word_vocab": 11827, "elapsed_s": 489.979807138443, "steps_per_sec": 1498.9229949888213}, {"sent": 42000, "train_loss": 2.0330393494128005, "train_acc": 0.4221956062659564, "eval_loss": 1.9251436953976735, "eval_acc": 0.4075829383886256, "ppl": 6.856133789990039, "w_freq": 0.00023821940146140168, "w_bigram": 0.0003441459331371583, "w_trigram": 0.030997428054289886, "w_skip_bi": 0.21924184189710616, "w_skip_tri": 0.6394717710576657, "w_neural_phoneme": 0.3237344301733989, "w_neural_word": 0.3237851543218467, "word_vocab": 12148, "elapsed_s": 511.2086036205292, "steps_per_sec": 1498.740425285816}, {"sent": 44000, "train_loss": 2.038187580584133, "train_acc": 0.4136109167249825, "eval_loss": 1.9579543555595058, "eval_acc": 0.4144144144144144, "ppl": 7.084819210230137, "w_freq": 2.3019951193833443e-05, "w_bigram": 3.326266031342418e-05, "w_trigram": 0.018392947942744674, "w_skip_bi": 0.19500459817758337, "w_skip_tri": 0.6527837747166838, "w_neural_phoneme": 0.32063096375579836, "w_neural_word": 0.3206122871697687, "word_vocab": 12373, "elapsed_s": 530.3895089626312, "steps_per_sec": 1498.4251886022773}, {"sent": 46000, "train_loss": 2.048150744103815, "train_acc": 0.41886444777150755, "eval_loss": 1.8764783324395902, "eval_acc": 0.41379310344827586, "ppl": 6.530466187797844, "w_freq": 0.00031922218074032855, "w_bigram": 0.0011571265157437353, "w_trigram": 0.030164394863674126, "w_skip_bi": 0.2140344615608487, "w_skip_tri": 0.6651480868640631, "w_neural_phoneme": 0.3172280283052047, "w_neural_word": 0.3172134653523036, "word_vocab": 12688, "elapsed_s": 553.4711921215057, "steps_per_sec": 1498.6886613204265}, {"sent": 48000, "train_loss": 1.9451634045900743, "train_acc": 0.4364364003461206, "eval_loss": 1.8530468554827837, "eval_acc": 0.453125, "ppl": 6.379226523734227, "w_freq": 0.00013785295253833406, "w_bigram": 0.001884357380645802, "w_trigram": 0.005010380926560289, "w_skip_bi": 0.22006032036827053, "w_skip_tri": 0.67502595, "w_neural_phoneme": 0.3147005045301676, "w_neural_word": 0.31466173150393156, "word_vocab": 12870, "elapsed_s": 572.0534970760345, "steps_per_sec": 1498.4909704800964}, {"sent": 50000, "train_loss": 1.911232161860392, "train_acc": 0.4444842663608343, "eval_loss": 1.7821394846204688, "eval_acc": 0.4444444444444444, "ppl": 5.942556836530168, "w_freq": 0.0, "w_bigram": 7.541563718851001e-05, "w_trigram": 0.00028189530155166477, "w_skip_bi": 0.22379146768347088, "w_skip_tri": 0.6844746126877431, "w_neural_phoneme": 0.3115181067964101, "w_neural_word": 0.31151843365851495, "word_vocab": 12991, "elapsed_s": 590.8203613758087, "steps_per_sec": 1498.118646315566}, {"sent": 52000, "train_loss": 2.1228203857356545, "train_acc": 0.40456334461784055, "eval_loss": 1.6271889526545587, "eval_acc": 0.4778761061946903, "ppl": 5.089547631552006, "w_freq": 0.0, "w_bigram": 0.002098976543544382, "w_trigram": 0.02271423506400377, "w_skip_bi": 0.12349239110741955, "w_skip_tri": 0.6975082, "w_neural_phoneme": 0.30785775977749996, "w_neural_word": 0.30785775977749996, "word_vocab": 13361, "elapsed_s": 614.9278926849365, "steps_per_sec": 1498.4716272613668}, {"sent": 54000, "train_loss": 1.9806389211921418, "train_acc": 0.42321141338555146, "eval_loss": 1.666916852962132, "eval_acc": 0.4631578947368421, "ppl": 5.295814825035445, "w_freq": 7.333336021799578e-05, "w_bigram": 0.00016144256122055113, "w_trigram": 0.019815972605825086, "w_skip_bi": 0.14794320041780842, "w_skip_tri": 0.7067238824233358, "w_neural_phoneme": 0.3051981776142868, "w_neural_word": 0.3051748411809417, "word_vocab": 13527, "elapsed_s": 634.1036429405212, "steps_per_sec": 1498.2566502761986}, {"sent": 56000, "train_loss": 1.9439236792217904, "train_acc": 0.45308924485125857, "eval_loss": 1.7826574565724627, "eval_acc": 0.45652173913043476, "ppl": 5.945635711611251, "w_freq": 1.6599741801710307e-05, "w_bigram": 0.0009983398137513443, "w_trigram": 0.06256862666891208, "w_skip_bi": 0.19073904878940245, "w_skip_tri": 0.7165292334125315, "w_neural_phoneme": 0.30245810878169044, "w_neural_word": 0.302450708570481, "word_vocab": 13709, "elapsed_s": 651.5659396648407, "steps_per_sec": 1497.6734365549544}, {"sent": 58000, "train_loss": 2.0386124324709827, "train_acc": 0.41543026706231456, "eval_loss": 1.761100588307048, "eval_acc": 0.45161290322580644, "ppl": 5.818838016613212, "w_freq": 0.00011122932428714702, "w_bigram": 9.952289526865809e-05, "w_trigram": 0.02460595268475821, "w_skip_bi": 0.14831385074101944, "w_skip_tri": 0.723821748582657, "w_neural_phoneme": 0.299729567685218, "w_neural_word": 0.2996310455683336, "word_vocab": 13922, "elapsed_s": 670.717779636383, "steps_per_sec": 1497.6164796832413}, {"sent": 60000, "train_loss": 2.0788808831051124, "train_acc": 0.4161963258078796, "eval_loss": 1.7260817254777665, "eval_acc": 0.46938775510204084, "ppl": 5.618595521559455, "w_freq": 0.000759915158142357, "w_bigram": 0.0009387574188833107, "w_trigram": 0.04035639989280677, "w_skip_bi": 0.11925771715744977, "w_skip_tri": 0.7371904045278395, "w_neural_phoneme": 0.2965167888325299, "w_neural_word": 0.2965136607424492, "word_vocab": 14280, "elapsed_s": 694.629296541214, "steps_per_sec": 1498.096906049884}, {"sent": 62000, "train_loss": 1.9769372195679191, "train_acc": 0.42850594283094856, "eval_loss": 1.3692545188855134, "eval_acc": 0.5689655172413793, "ppl": 3.93241805909645, "w_freq": 1.2353291739526551e-05, "w_bigram": 0.00020650629558154515, "w_trigram": 0.036212420452943594, "w_skip_bi": 0.12551715856908077, "w_skip_tri": 0.74990705, "w_neural_phoneme": 0.2929220231335842, "w_neural_word": 0.29292050194696206, "word_vocab": 14490, "elapsed_s": 715.058450460434, "steps_per_sec": 1498.0076094622284}, {"sent": 64000, "train_loss": 2.0677505325352192, "train_acc": 0.4212845459672871, "eval_loss": 1.8507718183601032, "eval_acc": 0.4153846153846154, "ppl": 6.364730042847392, "w_freq": 2.5957082866716556e-05, "w_bigram": 0.001565059352493228, "w_trigram": 0.03090386093705494, "w_skip_bi": 0.0932881134537389, "w_skip_tri": 0.7456391023314887, "w_neural_phoneme": 0.2901139538944109, "w_neural_word": 0.2900983213542502, "word_vocab": 14758, "elapsed_s": 734.2538552284241, "steps_per_sec": 1497.4807311810973}, {"sent": 66000, "train_loss": 2.0266912090326583, "train_acc": 0.4166428203513993, "eval_loss": 1.8057450735273728, "eval_acc": 0.40963855421686746, "ppl": 6.084503162314162, "w_freq": 2.103656501253577e-05, "w_bigram": 9.79229226875604e-05, "w_trigram": 0.03075416350244147, "w_skip_bi": 0.08247223338748873, "w_skip_tri": 0.7602349207840922, "w_neural_phoneme": 0.2865778665583748, "w_neural_word": 0.2865735492926551, "word_vocab": 15086, "elapsed_s": 757.8718175888062, "steps_per_sec": 1496.9246430212638}, {"sent": 68000, "train_loss": 2.0204846063983326, "train_acc": 0.41481168556851783, "eval_loss": 2.0005522425386477, "eval_acc": 0.4563106796116505, "ppl": 7.393137776963321, "w_freq": 6.548344654580944e-05, "w_bigram": 0.0012918031731307521, "w_trigram": 0.04641960829378828, "w_skip_bi": 0.0835012774971967, "w_skip_tri": 0.7821271630623516, "w_neural_phoneme": 0.2840051744647513, "w_neural_word": 0.2840335467551523, "word_vocab": 15342, "elapsed_s": 778.6454060077667, "steps_per_sec": 1496.5091824973497}, {"sent": 70000, "train_loss": 1.8987665856859937, "train_acc": 0.4526591270212169, "eval_loss": 1.6832346781773753, "eval_acc": 0.5194805194805194, "ppl": 5.382939918511319, "w_freq": 0.0, "w_bigram": 4.3070481957018897e-05, "w_trigram": 0.060077523614136916, "w_skip_bi": 0.11668519003713498, "w_skip_tri": 0.7939622282102138, "w_neural_phoneme": 0.2802699903068832, "w_neural_word": 0.2802665212371874, "word_vocab": 15567, "elapsed_s": 800.3778052330017, "steps_per_sec": 1495.9772649510624}, {"sent": 72000, "train_loss": 1.8113302657896218, "train_acc": 0.4669039145907473, "eval_loss": 1.901058476348171, "eval_acc": 0.4175824175824176, "ppl": 6.692975050079657, "w_freq": 1.724220892864327e-05, "w_bigram": 2.7731086590944795e-05, "w_trigram": 0.04063225720269605, "w_skip_bi": 0.13485883563040732, "w_skip_tri": 0.8018982802180007, "w_neural_phoneme": 0.2772610750853189, "w_neural_word": 0.2772626340840105, "word_vocab": 15738, "elapsed_s": 821.1420764923096, "steps_per_sec": 1495.7910880012046}, {"sent": 74000, "train_loss": 1.8558915220141143, "train_acc": 0.46303602799297294, "eval_loss": 1.7710075775706238, "eval_acc": 0.4666666666666667, "ppl": 5.876771682633129, "w_freq": 4.957693067452308e-06, "w_bigram": 0.0003516390248631074, "w_trigram": 0.03661104784800651, "w_skip_bi": 0.14040875594524663, "w_skip_tri": 0.817043, "w_neural_phoneme": 0.2739032619087128, "w_neural_word": 0.27396315004064303, "word_vocab": 15951, "elapsed_s": 844.3136689662933, "steps_per_sec": 1495.8658688379244}, {"sent": 76000, "train_loss": 1.7951495512194129, "train_acc": 0.48186675941442436, "eval_loss": 1.8820079949367525, "eval_acc": 0.48333333333333334, "ppl": 6.566677487628953, "w_freq": 8.643997017516979e-05, "w_bigram": 9.329529925191179e-05, "w_trigram": 0.041172590945193405, "w_skip_bi": 0.15507383800084015, "w_skip_tri": 0.82806642353532, "w_neural_phoneme": 0.2705774611194124, "w_neural_word": 0.27059365133736485, "word_vocab": 16111, "elapsed_s": 865.524932384491, "steps_per_sec": 1495.747784449609}, {"sent": 78000, "train_loss": 1.7885813137414621, "train_acc": 0.48145977797310935, "eval_loss": 1.7628009957335928, "eval_acc": 0.5154639175257731, "ppl": 5.828740829012325, "w_freq": 0.00010750517594666946, "w_bigram": 0.00031125295114216187, "w_trigram": 0.03379048878245142, "w_skip_bi": 0.15708075808273558, "w_skip_tri": 0.838573841826452, "w_neural_phoneme": 0.26763236875135904, "w_neural_word": 0.2676265455056914, "word_vocab": 16261, "elapsed_s": 886.062260389328, "steps_per_sec": 1495.7459077623555}, {"sent": 80000, "train_loss": 1.7484018847732679, "train_acc": 0.49511306086900425, "eval_loss": 1.6659690757310892, "eval_acc": 0.49074074074074076, "ppl": 5.290797950139903, "w_freq": 1.6476654909058993e-05, "w_bigram": 0.00018095676333165832, "w_trigram": 0.029356554991948693, "w_skip_bi": 0.16486658810482008, "w_skip_tri": 0.8496569372847096, "w_neural_phoneme": 0.2644817286686628, "w_neural_word": 0.2645014945752039, "word_vocab": 16375, "elapsed_s": 906.762975692749, "steps_per_sec": 1495.7878038235897}, {"sent": 82000, "train_loss": 1.8266525054997689, "train_acc": 0.4763773347983048, "eval_loss": 1.6947775549490514, "eval_acc": 0.4852941176470588, "ppl": 5.445434520518501, "w_freq": 4.758856451998734e-05, "w_bigram": 0.0, "w_trigram": 0.012052160465069923, "w_skip_bi": 0.14689155013615426, "w_skip_tri": 0.8604399388430413, "w_neural_phoneme": 0.26122889071792593, "w_neural_word": 0.2612134628951565, "word_vocab": 16520, "elapsed_s": 928.1313920021057, "steps_per_sec": 1495.6718541816658}, {"sent": 84000, "train_loss": 2.0140051565719834, "train_acc": 0.42844738474260907, "eval_loss": 1.6250920371957944, "eval_acc": 0.5084745762711864, "ppl": 5.078886462137244, "w_freq": 1.5155184047603248e-05, "w_bigram": 3.893904490734015e-05, "w_trigram": 4.464154619108076e-05, "w_skip_bi": 0.10301999077834527, "w_skip_tri": 0.8391374919767656, "w_neural_phoneme": 0.2563566031878798, "w_neural_word": 0.25635454417923537, "word_vocab": 16985, "elapsed_s": 959.9511001110077, "steps_per_sec": 1496.4824769031245}, {"sent": 86000, "train_loss": 2.065911053274546, "train_acc": 0.4153698520591763, "eval_loss": 1.6232812465001605, "eval_acc": 0.48333333333333334, "ppl": 5.069697983503955, "w_freq": 2.0213225077538084e-05, "w_bigram": 3.825364249725072e-05, "w_trigram": 0.0010616410868249104, "w_skip_bi": 0.07027956990203875, "w_skip_tri": 0.8300712402744733, "w_neural_phoneme": 0.25011190931691335, "w_neural_word": 0.2501106564582527, "word_vocab": 17701, "elapsed_s": 1000.7389311790466, "steps_per_sec": 1497.9681046622477}, {"sent": 88000, "train_loss": 1.999925072844319, "train_acc": 0.42861088097210714, "eval_loss": 1.8100833227976298, "eval_acc": 0.48936170212765956, "ppl": 6.110956593017489, "w_freq": 6.457073289841142e-05, "w_bigram": 0.004128396822787636, "w_trigram": 0.009727584032424392, "w_skip_bi": 0.06948713895303807, "w_skip_tri": 0.8482377605937955, "w_neural_phoneme": 0.23167771483085525, "w_neural_word": 0.2382937112514438, "word_vocab": 18150, "elapsed_s": 1031.6229994297028, "steps_per_sec": 1498.752936736322}, {"sent": 90000, "train_loss": 1.950716500312223, "train_acc": 0.43611584327086883, "eval_loss": 1.5672094590276084, "eval_acc": 0.4745762711864407, "ppl": 4.79325374262496, "w_freq": 2.761688049160913e-05, "w_bigram": 0.005022580590965811, "w_trigram": 0.016042329552083585, "w_skip_bi": 0.07393772473331248, "w_skip_tri": 0.865957684143416, "w_neural_phoneme": 0.2297602805479303, "w_neural_word": 0.2384797736164583, "word_vocab": 18315, "elapsed_s": 1050.490453004837, "steps_per_sec": 1498.6561710263834}, {"sent": 92000, "train_loss": 1.9736854264747719, "train_acc": 0.436160982446773, "eval_loss": 1.7921611888485354, "eval_acc": 0.5535714285714286, "ppl": 6.002410801923678, "w_freq": 8.341029986649858e-06, "w_bigram": 0.0004170255873509261, "w_trigram": 0.015168363797333564, "w_skip_bi": 0.06802639055348417, "w_skip_tri": 0.8721088054765493, "w_neural_phoneme": 0.22465665838377202, "w_neural_word": 0.2370220472918633, "word_vocab": 18506, "elapsed_s": 1070.5428359508514, "steps_per_sec": 1498.5761859543684}, {"sent": 94000, "train_loss": 1.9733076789695052, "train_acc": 0.4435667471541911, "eval_loss": 1.8381862074137196, "eval_acc": 0.41353383458646614, "ppl": 6.285127997663289, "w_freq": 1.7201596030863343e-05, "w_bigram": 0.0001805887476969183, "w_trigram": 0.01587563469626517, "w_skip_bi": 0.06580874734118058, "w_skip_tri": 0.8759190804054925, "w_neural_phoneme": 0.22291132574437858, "w_neural_word": 0.23672067937371336, "word_vocab": 18678, "elapsed_s": 1089.9353623390198, "steps_per_sec": 1498.510880952567}, {"sent": 96000, "train_loss": 1.9541992160740274, "train_acc": 0.4419679759973058, "eval_loss": 1.9098555305630687, "eval_acc": 0.38562091503267976, "ppl": 6.752113254064892, "w_freq": 4.1234338054432075e-05, "w_bigram": 0.0006160683321713935, "w_trigram": 0.018357954026768385, "w_skip_bi": 0.06464110204488131, "w_skip_tri": 0.8816573730505688, "w_neural_phoneme": 0.22164842275839133, "w_neural_word": 0.2369715209000223, "word_vocab": 18873, "elapsed_s": 1112.4942667484283, "steps_per_sec": 1497.4845711961989}, {"sent": 98000, "train_loss": 1.92013704162962, "train_acc": 0.4429237429237429, "eval_loss": 1.884457112654517, "eval_acc": 0.4387755102040816, "ppl": 6.582779763947173, "w_freq": 2.5116430657434587e-06, "w_bigram": 0.002147201247069049, "w_trigram": 0.022342940007730933, "w_skip_bi": 0.06639164807864611, "w_skip_tri": 0.8901679424705264, "w_neural_phoneme": 0.22099201709742008, "w_neural_word": 0.23756999875233156, "word_vocab": 19035, "elapsed_s": 1132.5879561901093, "steps_per_sec": 1497.4316040804908}, {"sent": 100000, "train_loss": 1.9026785745815122, "train_acc": 0.4583273355405211, "eval_loss": 1.8736170865306048, "eval_acc": 0.4322033898305085, "ppl": 6.5118076242329535, "w_freq": 6.303877729132668e-05, "w_bigram": 0.0036132175251367142, "w_trigram": 0.024671714119590894, "w_skip_bi": 0.06795049134601011, "w_skip_tri": 0.8981083792823219, "w_neural_phoneme": 0.22149772291835784, "w_neural_word": 0.23874823213661373, "word_vocab": 19189, "elapsed_s": 1151.2511730194092, "steps_per_sec": 1497.2935884000517}, {"sent": 102000, "train_loss": 1.9970472967983357, "train_acc": 0.44037114845938374, "eval_loss": 1.9204054272732058, "eval_acc": 0.4177215189873418, "ppl": 6.8237244325441875, "w_freq": 3.182927799963724e-05, "w_bigram": 3.707779085204988e-05, "w_trigram": 0.021915685177146982, "w_skip_bi": 0.06117124107125588, "w_skip_tri": 0.8959566290270812, "w_neural_phoneme": 0.22096258852157363, "w_neural_word": 0.23943868228455045, "word_vocab": 19394, "elapsed_s": 1170.4331676959991, "steps_per_sec": 1497.1559661534957}, {"sent": 104000, "train_loss": 1.9732259350545995, "train_acc": 0.4385173745173745, "eval_loss": 1.6547392372134018, "eval_acc": 0.5230769230769231, "ppl": 5.231715507577078, "w_freq": 2.2971012109444218e-06, "w_bigram": 3.1178074941946244e-06, "w_trigram": 0.02200624828058436, "w_skip_bi": 0.05939776895455238, "w_skip_tri": 0.8977385033708655, "w_neural_phoneme": 0.2207694209870391, "w_neural_word": 0.23991041117215486, "word_vocab": 19634, "elapsed_s": 1192.020720243454, "steps_per_sec": 1497.2021624217239}, {"sent": 106000, "train_loss": 1.9455210661806934, "train_acc": 0.4422370390529324, "eval_loss": 1.9339221057810974, "eval_acc": 0.42105263157894735, "ppl": 6.916584689245599, "w_freq": 1.3319510177449202e-05, "w_bigram": 0.0017174336648858527, "w_trigram": 0.02399738528062805, "w_skip_bi": 0.05949208490244423, "w_skip_tri": 0.9008616615015612, "w_neural_phoneme": 0.2214823828940293, "w_neural_word": 0.2408385997163679, "word_vocab": 19837, "elapsed_s": 1211.6544728279114, "steps_per_sec": 1497.2024126367019}, {"sent": 108000, "train_loss": 1.8134665592328145, "train_acc": 0.47753083404567104, "eval_loss": 2.0303747976539177, "eval_acc": 0.4166666666666667, "ppl": 7.616940635338638, "w_freq": 0.00023797268771405183, "w_bigram": 0.003059440443657493, "w_trigram": 0.027653378154983796, "w_skip_bi": 0.06336487449287119, "w_skip_tri": 0.9088894449869791, "w_neural_phoneme": 0.2223456029895363, "w_neural_word": 0.2419230736155544, "word_vocab": 20002, "elapsed_s": 1234.0368134975433, "steps_per_sec": 1496.590684977711}, {"sent": 110000, "train_loss": 1.7733889190741998, "train_acc": 0.48939571913338553, "eval_loss": 1.788419834094627, "eval_acc": 0.44554455445544555, "ppl": 5.979995611471487, "w_freq": 1.320926421354793e-06, "w_bigram": 0.0025843213052118437, "w_trigram": 0.029221644294984764, "w_skip_bi": 0.0663518737527829, "w_skip_tri": 0.9162281500611833, "w_neural_phoneme": 0.22304740879811794, "w_neural_word": 0.24289822394487073, "word_vocab": 20143, "elapsed_s": 1254.5816235542297, "steps_per_sec": 1496.511637625501}, {"sent": 112000, "train_loss": 1.7932382043835773, "train_acc": 0.48665634674922603, "eval_loss": 1.843002322908089, "eval_acc": 0.44166666666666665, "ppl": 6.315470908829018, "w_freq": 5.11754185171025e-06, "w_bigram": 0.002387639790171045, "w_trigram": 0.030881657173619674, "w_skip_bi": 0.06816132448090961, "w_skip_tri": 0.9219253930601433, "w_neural_phoneme": 0.2237189384447766, "w_neural_word": 0.2438901824805881, "word_vocab": 20327, "elapsed_s": 1276.1286475658417, "steps_per_sec": 1496.5544450732693}, {"sent": 114000, "train_loss": 1.7266746649828377, "train_acc": 0.5010028078620137, "eval_loss": 1.794911364547625, "eval_acc": 0.4411764705882353, "ppl": 6.018941206586753, "w_freq": 8.036493813362989e-06, "w_bigram": 0.00305448920895356, "w_trigram": 0.03269685440654975, "w_skip_bi": 0.07045248088573455, "w_skip_tri": 0.9262209895060376, "w_neural_phoneme": 0.2242987091750386, "w_neural_word": 0.244488353199263, "word_vocab": 20431, "elapsed_s": 1296.3175764083862, "steps_per_sec": 1496.324693347305}, {"sent": 116000, "train_loss": 1.7718260174127611, "train_acc": 0.4940725819766866, "eval_loss": 1.8521200675469136, "eval_acc": 0.42105263157894735, "ppl": 6.373317072378946, "w_freq": 1.861360303745382e-06, "w_bigram": 0.0034358019162241404, "w_trigram": 0.033914831474538186, "w_skip_bi": 0.07189891359405522, "w_skip_tri": 0.9295183452842231, "w_neural_phoneme": 0.2248534572205537, "w_neural_word": 0.24508020972288558, "word_vocab": 20579, "elapsed_s": 1316.8358359336853, "steps_per_sec": 1496.0065227896846}, {"sent": 118000, "train_loss": 1.7382876405514467, "train_acc": 0.49921080876317947, "eval_loss": 1.8135424949270964, "eval_acc": 0.4424778761061947, "ppl": 6.1321320474040295, "w_freq": 1.526479999066242e-05, "w_bigram": 0.0031323819240634404, "w_trigram": 0.03429316805648019, "w_skip_bi": 0.07272670307703355, "w_skip_tri": 0.9325042877002102, "w_neural_phoneme": 0.2253259982643581, "w_neural_word": 0.2456475473441679, "word_vocab": 20697, "elapsed_s": 1338.2457747459412, "steps_per_sec": 1495.7439341663583}, {"sent": 120000, "train_loss": 1.7551960944586795, "train_acc": 0.4923213824382122, "eval_loss": 1.9342081539625138, "eval_acc": 0.4224137931034483, "ppl": 6.918563448714351, "w_freq": 8.027249703507432e-06, "w_bigram": 0.002837941605794471, "w_trigram": 0.034665651595214515, "w_skip_bi": 0.07333538087908563, "w_skip_tri": 0.9357541672350208, "w_neural_phoneme": 0.22597594249365366, "w_neural_word": 0.24634007498400295, "word_vocab": 20840, "elapsed_s": 1361.5818145275116, "steps_per_sec": 1495.6941832442865}, {"sent": 122000, "train_loss": 1.7341605417912884, "train_acc": 0.5040394358482815, "eval_loss": 1.6758216454949606, "eval_acc": 0.47297297297297297, "ppl": 5.343183548663422, "w_freq": 2.414910602071845e-06, "w_bigram": 0.003156939996273343, "w_trigram": 0.03570601222314808, "w_skip_bi": 0.07453643245233657, "w_skip_tri": 0.9387213952996164, "w_neural_phoneme": 0.2266202392903855, "w_neural_word": 0.24697003235638615, "word_vocab": 20947, "elapsed_s": 1381.526701450348, "steps_per_sec": 1495.2458014972663}, {"sent": 124000, "train_loss": 1.7326503312113062, "train_acc": 0.5012925530258199, "eval_loss": 2.0732758468460153, "eval_acc": 0.38613861386138615, "ppl": 7.950826192456674, "w_freq": 8.257859966245066e-07, "w_bigram": 0.003224858622212897, "w_trigram": 0.0362407685020796, "w_skip_bi": 0.0752376856437425, "w_skip_tri": 0.9404047055409357, "w_neural_phoneme": 0.22697156690818554, "w_neural_word": 0.2473151031248518, "word_vocab": 21074, "elapsed_s": 1403.1753098964691, "steps_per_sec": 1495.0583759592982}, {"sent": 126000, "train_loss": 1.7836909718143563, "train_acc": 0.48380182497397267, "eval_loss": 1.6762865032231629, "eval_acc": 0.4854368932038835, "ppl": 5.345667946230187, "w_freq": 2.3948590077266986e-06, "w_bigram": 0.0032665722541187557, "w_trigram": 0.036402651612389536, "w_skip_bi": 0.07526309404718656, "w_skip_tri": 0.9414502621474398, "w_neural_phoneme": 0.22732979087012206, "w_neural_word": 0.24766674817640427, "word_vocab": 21232, "elapsed_s": 1425.1720685958862, "steps_per_sec": 1494.8980877088104}, {"sent": 128000, "train_loss": 1.9156972617702686, "train_acc": 0.4555644327474387, "eval_loss": 1.9520231917025068, "eval_acc": 0.43820224719101125, "ppl": 7.042922357967988, "w_freq": 4.241549340590268e-06, "w_bigram": 0.0024209983050539324, "w_trigram": 0.03554531285154458, "w_skip_bi": 0.07371224759247018, "w_skip_tri": 0.9410956502514535, "w_neural_phoneme": 0.22772106428854766, "w_neural_word": 0.24808833063110158, "word_vocab": 21541, "elapsed_s": 1453.6073627471924, "steps_per_sec": 1494.9312006050495}, {"sent": 130000, "train_loss": 2.0269157785974343, "train_acc": 0.4253824362606232, "eval_loss": 2.141725163840275, "eval_acc": 0.3392857142857143, "ppl": 8.514113206594372, "w_freq": 9.99786606573586e-06, "w_bigram": 0.0009099263946861115, "w_trigram": 0.03411228001428117, "w_skip_bi": 0.07063848627312498, "w_skip_tri": 0.9390616437802706, "w_neural_phoneme": 0.22814623468035108, "w_neural_word": 0.2485795453326677, "word_vocab": 22065, "elapsed_s": 1488.767904996872, "steps_per_sec": 1495.1914213953162}, {"sent": 132000, "train_loss": 2.03390155325021, "train_acc": 0.4314489928525016, "eval_loss": 1.8316893904802116, "eval_acc": 0.41346153846153844, "ppl": 6.244427028193549, "w_freq": 1.8078212845202246e-06, "w_bigram": 3.285060132663803e-07, "w_trigram": 0.03302047957985557, "w_skip_bi": 0.06799104790016235, "w_skip_tri": 0.9373729741297487, "w_neural_phoneme": 0.22859254148783387, "w_neural_word": 0.24905295751129988, "word_vocab": 22508, "elapsed_s": 1520.7646691799164, "steps_per_sec": 1495.1044340253582}, {"sent": 134000, "train_loss": 2.017445611076862, "train_acc": 0.4312743628185907, "eval_loss": 2.0458193964993394, "eval_acc": 0.4, "ppl": 7.735494378787105, "w_freq": 6.09112510160719e-07, "w_bigram": 1.932988274510006e-06, "w_trigram": 0.03193802884615008, "w_skip_bi": 0.06602486218254995, "w_skip_tri": 0.936068570213186, "w_neural_phoneme": 0.22884851840613388, "w_neural_word": 0.2493547852949733, "word_vocab": 23214, "elapsed_s": 1564.3069865703583, "steps_per_sec": 1496.127051846249}, {"sent": 136000, "train_loss": 1.9177536909888693, "train_acc": 0.453125, "eval_loss": 1.93398566794929, "eval_acc": 0.3565217391304348, "ppl": 6.9170243363372474, "w_freq": 1.4237730032719664e-05, "w_bigram": 4.773753514077895e-05, "w_trigram": 0.03181845876097105, "w_skip_bi": 0.06564865840616795, "w_skip_tri": 0.9362292314158513, "w_neural_phoneme": 0.22904496729543875, "w_neural_word": 0.24954668190821208, "word_vocab": 23433, "elapsed_s": 1587.915765762329, "steps_per_sec": 1496.3325204213852}, {"sent": 138000, "train_loss": 1.9817818033428023, "train_acc": 0.4286567362239989, "eval_loss": 1.5822435181680037, "eval_acc": 0.5277777777777778, "ppl": 4.865860220350948, "w_freq": 4.981242287851702e-05, "w_bigram": 5.506247584453079e-05, "w_trigram": 0.03169219087629866, "w_skip_bi": 0.06517248491365542, "w_skip_tri": 0.9360707273439266, "w_neural_phoneme": 0.22922453316667146, "w_neural_word": 0.2497157485013631, "word_vocab": 23628, "elapsed_s": 1608.0757825374603, "steps_per_sec": 1496.3181624457723}, {"sent": 140000, "train_loss": 2.134091683552672, "train_acc": 0.40918439488399, "eval_loss": 1.933672992490033, "eval_acc": 0.4536082474226804, "ppl": 6.914861890665657, "w_freq": 0.00019372669404656148, "w_bigram": 4.373788026974246e-05, "w_trigram": 0.03131083098385021, "w_skip_bi": 0.06398416422855717, "w_skip_tri": 0.9351400043277288, "w_neural_phoneme": 0.22942477406922024, "w_neural_word": 0.24989767218449874, "word_vocab": 23923, "elapsed_s": 1628.8552639484406, "steps_per_sec": 1496.3336853464903}, {"sent": 142000, "train_loss": 1.8778892486050764, "train_acc": 0.4605023537092396, "eval_loss": 2.149309841006978, "eval_acc": 0.375, "ppl": 8.578935524048436, "w_freq": 0.00017272487494976468, "w_bigram": 0.00012531810158663846, "w_trigram": 0.03143425851919989, "w_skip_bi": 0.06395850059674751, "w_skip_tri": 0.9354906189504653, "w_neural_phoneme": 0.22960752083070338, "w_neural_word": 0.2500655294432953, "word_vocab": 24057, "elapsed_s": 1648.3191742897034, "steps_per_sec": 1496.3206389095312}, {"sent": 144000, "train_loss": 1.9853750605424931, "train_acc": 0.4362357336758224, "eval_loss": 2.323899408934437, "eval_acc": 0.45714285714285713, "ppl": 10.215430884578893, "w_freq": 0.00014701870050829791, "w_bigram": 0.00016228382813346728, "w_trigram": 0.03137306280770583, "w_skip_bi": 0.06368451125728934, "w_skip_tri": 0.9354044183409257, "w_neural_phoneme": 0.22971334751170855, "w_neural_word": 0.25016342872253133, "word_vocab": 24250, "elapsed_s": 1671.202244758606, "steps_per_sec": 1496.3317622644802}, {"sent": 146000, "train_loss": 1.9211976908972084, "train_acc": 0.4553778613985575, "eval_loss": 1.7527877024671192, "eval_acc": 0.453125, "ppl": 5.770667177204251, "w_freq": 0.00014761854723064908, "w_bigram": 0.00018563874604355857, "w_trigram": 0.03136339818583566, "w_skip_bi": 0.06353485016878424, "w_skip_tri": 0.9355588368542496, "w_neural_phoneme": 0.2298147519862598, "w_neural_word": 0.2502559937468263, "word_vocab": 24421, "elapsed_s": 1694.930011510849, "steps_per_sec": 1494.1991603195997}, {"sent": 148000, "train_loss": 1.8726334683421944, "train_acc": 0.4569213582044077, "eval_loss": 2.0006443871353277, "eval_acc": 0.38235294117647056, "ppl": 7.393819046049129, "w_freq": 0.00011870151530824642, "w_bigram": 0.000262460235786491, "w_trigram": 0.031428301365247845, "w_skip_bi": 0.0635331991245312, "w_skip_tri": 0.9357451736107169, "w_neural_phoneme": 0.22990826862907196, "w_neural_word": 0.25034149654324866, "word_vocab": 24532, "elapsed_s": 1715.380619764328, "steps_per_sec": 1493.605541813805}, {"sent": 150000, "train_loss": 2.0038219784325926, "train_acc": 0.43624468833114244, "eval_loss": 1.9970044540164273, "eval_acc": 0.4117647058823529, "ppl": 7.366954960625814, "w_freq": 0.00013266909661891082, "w_bigram": 0.00027706064913079944, "w_trigram": 0.03136897483311752, "w_skip_bi": 0.06322694723819182, "w_skip_tri": 0.9356360315167365, "w_neural_phoneme": 0.2300129137473327, "w_neural_word": 0.2504367173998546, "word_vocab": 24715, "elapsed_s": 1737.7841699123383, "steps_per_sec": 1493.173343920433}], "mixing_weights": {"w_freq": 0.00013266909661891082, "w_bigram": 0.00027706064913079944, "w_trigram": 0.03136897483311752, "w_skip_bi": 0.06322694723819182, "w_skip_tri": 0.9356360315167365, "w_neural_phoneme": 0.2300129137473327, "w_neural_word": 0.2504367173998546}, "word_vocab": 24715, "elapsed_s": 1737.7841699123383, "steps_per_sec": 1493.173343920433}